Naver News & Blog Scraper - Cafe Keyword Monitor (Official API)

A Naver News scraper and Naver Blog scraper for keyword monitoring in Korea: it collects Korean news (네이버 뉴스), Naver Blog posts, Naver Cafe posts and Korean web documents for your keywords through the official Naver Search API on NAVER API HUB (Naver Cloud Platform), so there is no page scraping and no blocking. You run it with your own API key; it is automation software for your own searches, not a data feed. It is built for alerts and media monitoring: "only new since last run" mode, a date cutoff, publisher names in English, and de-duplication across keywords.

Live-tested on Apify: YES (2026-10-29). Keyword "삼성전자" on news, blog, cafe and web documents: 400 items in 7 s through NAVER API HUB keys. Publisher domain is always filled; an English publisher name is filled only for outlets that have an established English name (about 1 in 7 in that run).

Why this actor

• Official API: no blocking and no proxies, and it costs far less than headless scraping.
• NAVER API HUB ready: uses the new Naver Cloud Platform endpoints first and still accepts old NAVER Developers keys as a fallback until they stop working on 2027-06-30.
• Alert mode (onlyNewItems): remembers the links it already returned (ids only, kept less than 21 days), so a scheduled run only outputs new items, and you only pay for new items.
• English publisher names for about 50 major Korean outlets (Yonhap, Chosun, JoongAng, Hankyung, Maeil Business, KBS and others), plus the publisher domain.
• Privacy by design: blogger nicknames and profile links are dropped. Cafe posts keep only the community name.

Input

{
  "keywords": ["삼성전자", "\"K-뷰티\" -광고"],
  "sources": ["news", "blog", "cafearticle"],
  "sort": "date", "maxItemsPerSource": 100, "publishedAfter": "2026-10-01", "onlyNewItems": true,
  "apiProvider": "auto",
  "apiHubClientId": "YOUR_API_HUB_CLIENT_ID", "apiHubClientSecret": "YOUR_API_HUB_CLIENT_SECRET"
}

Sources: news, blog, cafearticle, webkr (web documents). Sort: date or sim (relevance). Optional fallback: legacyClientId / legacyClientSecret (an existing developers.naver.com app). apiProvider: auto (API HUB, then legacy on 401/403), apihub or legacy.

Output

Example (illustrative values), one news item:

{
  "keyword": "삼성전자", "source": "news", "rank": 1,
  "title": "삼성전자, 3분기 영업이익 \"시장 기대 상회\"", "snippet": "삼성전자가 28일 잠정 실적을 발표했다...",
  "url": "https://n.news.naver.com/mnews/article/001/0015000000",
  "originalUrl": "https://www.yna.co.kr/view/AKR20261028000100003",
  "publisherDomain": "yna.co.kr", "publisherNameEn": "Yonhap News Agency",
  "cafeName": null, "publishedAt": "2026-10-28T00:12:00.000Z", "fetchedAt": "2026-10-29T06:00:00.000Z"
}

Blog items have publisherDomain: "blog.naver.com" and a day-only publishedAt ("2026-10-27"). Cafe items have cafeName and no date.

Endpoints used (official)

Primary, NAVER API HUB (Naver Cloud Platform): GET https://naverapihub.apigw.ntruss.com/search/v1/{news|blog|cafearticle|webkr} with headers X-NCP-APIGW-API-KEY-ID and X-NCP-APIGW-API-KEY.

Fallback until 2027-06-30, legacy NAVER Developers keys: GET https://openapi.naver.com/v1/search/{news|blog|cafearticle|webkr}.json with headers X-Naver-Client-Id and X-Naver-Client-Secret.

Getting API HUB keys: Naver Cloud Platform console > Services > Application Services > NAVER API HUB > Application > register an application and add the Search APIs (news, blog, cafe, web documents) > copy the Client ID and Secret from 인증 정보. API HUB currently lists a free tier for the Search API (25,000 calls per day); Naver says the free policy may change. A Naver Cloud Platform account (with a payment method) is required.

API keys (secret inputs)

• All four credential fields are secret inputs (isSecret in the input schema): Apify stores them encrypted and hides them in the Console and in API responses.
• The actor sends them only as request headers to the two official Naver hosts above. They are never logged, never written to the dataset or key-value stores, and never put in URLs.
• If a key is exposed, reissue it in the provider console (API HUB: 인증 정보 > 재발급) and update the input.

Terms notice (read before use)

This actor is a tool you run with your own NAVER key. You are the API user under the NAVER API terms (developers.naver.com, 이용약관) and the NAVER API HUB / Naver Cloud Platform terms, and you are responsible for how you use the results. Among other things, Naver's search API special terms (as revised 2026-10-07) say:

• results may be stored on a server only for history lookup and for at most 21 days; no other copying, storing or caching;
• no providing or selling the results to third parties;
• no feeding the results into AI for training, improvement, evaluation or display;
• no earning revenue from the search API service itself.

What the actor does to stay inside these rules: the onlyNewItems memory keeps only item ids and the date first seen, and drops them before 21 days; it keeps no titles or snippets between runs. Run datasets live in your Apify account: delete them (or rely on your plan's data retention) within 21 days, and do not copy them into long-lived storage. The developer of this actor does not see, keep or resell your results.

Limitations

• The Naver API returns at most 1,000 items per keyword per source, and only a title and snippet, not the full article text. Fetching full article text is out of scope on purpose (copyright).
• Blog dates are day-only, and cafe posts have no date in the API, so publishedAfter is not applied to them. Web documents have no date at all.
• Snippets are Naver's, in Korean. No machine translation is included (sending them to an AI translator would conflict with the terms above).
• If the API HUB free tier ends, calls will be billed to your Naver Cloud account by Naver, not by this actor.
• Because onlyNewItems forgets ids after less than 21 days, a very old item that Naver re-surfaces can appear again.

Pricing

• Free from 12 November 2026. Until then the Store shows the earlier pay-per-result price, which is being removed.
• You pay only Apify's normal platform usage for the run (a fraction of a cent for a typical run).
• Naver data is fetched with your own key; Naver does not charge for it on the current free tier.
• Why free: Naver's API terms do not allow earning revenue from the API itself, so this actor is offered at no charge.

FAQ

Do I need my own API key? Yes. Register an application in the NAVER API HUB (Naver Cloud Platform console), add the Search APIs, and paste the Client ID and Secret into the input. The steps are under "Endpoints used" above. Existing NAVER Developers keys also work until 2027-06-30.

How fresh is the data? Each run queries the Naver Search API live, so you get whatever Naver's search index holds at that moment. Nothing is cached between runs except the item ids used by onlyNewItems.

Can I schedule it daily? Yes. Create an Apify Schedule (daily or hourly) with onlyNewItems: true and sort: "date". Each run then returns only items it has not returned in the last 21 days, so you pay only for new items.

What does a typical run cost? It is free (from 12 November 2026); you only pay Apify platform usage, usually well under one cent per run.

Is this legal, and where does the data come from? The data comes from Naver's official Search API, called with your key under Naver's API terms. No pages are scraped. Read the terms notice above: it limits storage to 21 days and forbids reselling the results or feeding them into AI.